Research On Indoor Path Perception Technology Of Nursing Robot

Perceptive robot is an important part of the whole robot field.The nursing robot independently designed and developed by our research group has a complex mechanical structure and works in an indoor environment.shapes and positions,mutual shielding between objects and other issues will affect the robot's perception of the environment,thus affecting the work of the nursing robot.Therefore,aiming at the complexity of nursing robot structure and the diversity of indoor environment,this paper proposes hierarchical environment perception technology for nursing robot,and proposes A piecewise planning A* algorithm optimized by cost function for hierarchical perception to plan the perception path.The main research contents of this paper are as follows:According to the structure characteristics of the nursing robot developed by our laboratory,the selection and design of the perceptual equipment were completed.After the research and simulation comparison of multiple path planning algorithms,the traditional A* algorithm has many problems,such as multiple turning points and paths close to obstacles.A cost function optimization A* algorithm is proposed.The cost estimation function of the algorithm proposed in this paper can effectively reduce the number of turning points.On the basis of the optimized algorithm,the segmented planning and adding virtual obstacle points are carried out.After optimization,the algorithm can reduce the number of turns,and the segmented planning conforms to the path planning method of the nursing robot.The virtual obstacle point can keep a safe distance between the nursing robot and the obstacle.A calibrated environment was set up in the laboratory for the distance perception experiment of the lidar,and the experimental results showed that the lidar imaging effect was good in the range of 1m to 9.9m.The imaging effect is unstable if the distance is less than 1m.Radar imaging at places over 9.9m gradually became unclear,and imaging at places over 9.98 m and thereafter was impossible.Within the range of 1m-9.9m,the imaging noise increases with the increase of the distance.The high-rise obstacles were built and the nursing robot was simulated to conduct the layered perception experiment of the obstacles in the indoor environment.Through analysis,the lidar can clearly perceive the environmental information at all levels and the change of obstacles,so as to obtain the longitudinal change of obstacles.Finally,the actual environment and physical object perception in the laboratory are conducted.The experimental results show that the laser radar has a good imaging effect on rectangular objects with regular shape and large volume,and the perception effect of obstacles with other shapes is unstable.Carry out Matlab simulation on the optimized and improved A* algorithm,and compare and analyze the data with the traditional A* algorithm.The analysis results of simulation experimental data verify that the optimized A* algorithm can effectively reduce the number of turns by 50-80%.The path planned by the new algorithm can maintain A relatively safe distance from obstacles at both straight and turning points.According to the result of perception experiment,the algorithm is applied to the hierarchical perception environment map of nursing robot.The layered simulation of road environment and obstacles was carried out by Matlab,and the path planning was carried out.The application results show that the algorithm effectively realizes the path planning problem after the hierarchical environment perception of nursing robot.